Back to AI Radar

DeepSeek R1

Planning/Reasoning / 2025 / core

Summary

Henry's source note frames this paper around Building OSS o1-level reasoning model with pure RL, no SFT, no RM.

Why George Should Read It

Henry marked this as a priority paper for catching up to the AI research frontier.

Best For

  • Planning/Reasoning
  • AI research catch-up
  • builder reading path

Notes

  • Henry note: Building OSS o1-level reasoning model with pure RL, no SFT, no RM

Next Reading Action

Read the abstract and skim the figures before deciding whether to take deeper notes.

Copyable Markdown

# DeepSeek R1

Year: 2025
Category: Planning/Reasoning

Why it matters:
Henry marked this as a priority paper for catching up to the AI research frontier.

Best for:
- Planning/Reasoning
- AI research catch-up
- builder reading path

Next step:
Read the abstract and skim the figures before deciding whether to take deeper notes.